Fundus reflectometry, together with a model fit procedure, provided information on a set of parameters: lens optical density, macular pigment optical density, melanin, blood, the directional cone reflectance (Rd), a measure for foveal cone photoreceptor integrity, and RILM, the reflectance at the inner limiting membrane.
Model fit procedures for both linear and logistic models provided no indication of a violation of basic assumptions of the respective models and both procedures accounted for reasonable amounts of variation of outcome variables.
To this purpose a general model fitting procedure has been developed.
Moreover, in a rigorous and computationally intensive move, an outer loop of nested cross-validation was used as an unbiased method to evaluate overall performance of the internally cross-validated model fitting procedure.
Then, a statistical formulation of general nonlinear structural equation analysis is introduced, and a general model fitting procedure applicable under weak assumptions on latent variable distributions is developed.
Results based on extensive testing and evaluation, show that the new procedure performs consistently well in fitting a set of training data as well as predicting a set of validation data, and outperforms a calibration procedure and an empirical model fitting procedure.
We tested the lag time between temperature, rainfall, and HFMD by including lag terms 1 2 and 3 4 weeks using a backward stepwise model fitting procedure.
We developed a maximum likelihood model fitting procedure to a collection of HIV-1 alignments sampled from different viral genes, and inferred two empirical substitution models, suitable for describing between-and within-host evolution.
As noted above, an important application of the Bayesian model fitting procedure we use is to produce better estimates of the parameters underlying LF transmission.
Our model fitting procedure is iterative in order to minimize confounding between the biological variation of interest and the other sources of variation.
Although the model fitting procedure we use is somewhat robust to these outliers, they can still affect the posterior distribution on the ordering, which manifests as uncertainty.
